A service recommendation method includes, when a user of a terminal requests a first service from an intelligent assistant, selecting, according to a name of the first service and by using a pre-established service relationship model, a potential service with a degree of relevance to the first service that meets a preset condition from multiple services that the intelligent assistant can provide, where names of the multiple services and degrees of relevance of the multiple services to each other are recorded in the service relationship model; and recommending the potential service to the user.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A service recommendation method, comprising: receiving, by an intelligent assistant, a request for a first service; parsing, by the intelligent assistant, the request for the first service using natural language processing; extracting, by the intelligent assistant, a service name from the request after parsing the request, wherein the service name indicates a type of the first service; determining a type of service parameter for the first service based on the service name, wherein the type of service parameter for the first service is recorded in a pre-established service relationship model indicating a relationship between two or more services, wherein the pre-established service relationship model comprises a plurality of service names of a plurality of services provided by the intelligent assistant, corresponding degrees of relevance of the plurality of services to the first service, a plurality of corresponding trigger conditions of the plurality of services, and types of service parameters of the plurality of services, wherein each service parameter is a configuration parameter of a third-party application that provides a corresponding service, wherein each service parameter comprises a mandatory parameter and an optional parameter, wherein the mandatory parameter is a parameter that must be received to invoke the first potential service, and wherein the optional parameter is a parameter that is optionally received with the mandatory parameter; extracting, by the intelligent assistant, a service parameter for the first service from the request based on the type of service parameter indicated in the pre-established service relationship model after parsing the request; selecting, by the intelligent assistant, a first potential service with a first degree of relevance to the first service based on the service name, the service parameter, and the pre-established service relationship model indicating that a degree of relevance between the first potential service and the first service meets a threshold; generating, by the intelligent assistant, an instruction for requesting one of the third-party applications to provide the first potential service; detecting, by the intelligent assistant, whether a corresponding trigger condition for performing the first potential service, as indicated in the pre-established service relationship model, is met after generating the instruction; and determining, by the intelligent assistant, whether to store the instruction in a cache temporarily or send the instruction to the one of the third-party applications based on whether the corresponding trigger condition is met.
2. The service recommendation method of claim 1 , wherein the threshold comprises at least one of a relevance degree or a quantity of recommended services.
3. The service recommendation method of claim 1 , wherein the threshold is for a quantity of services recommended to the user, and wherein selecting the first potential service comprises: ranking the corresponding degrees of relevance of the plurality of services to the first service; and selecting the first potential service corresponding to a ranking that meets the threshold.
4. The service recommendation method of claim 1 , further comprising selecting, from the pre-established service relationship model, a second potential service with a second degree of relevance to the first service that meets the corresponding trigger condition, wherein the first degree of relevance is higher than the second degree of relevance, and wherein recommending the first potential service comprises recommending, according to a ranking of the first degree of relevance and the second degree of relevance, the first potential service.
5. The service recommendation method of claim 1 , wherein the pre-established service relationship model further comprises corresponding relevance conditions of the plurality of services, and wherein selecting the first potential service comprises: determining a first relevance condition based on the request for the first service and an environmental condition of the first request; and selecting the first potential service when the first relevance condition meets the corresponding trigger condition.
6. The service recommendation method of claim 1 , further comprising: extracting, from the pre-established service relationship model, a first of the corresponding service parameters corresponding to the first service; extracting, from the pre-established service relationship model, a second of the corresponding service parameters corresponding to the first potential service; allocating, in a session cache, a first session to the first service; storing the first service parameter in the first session; allocating, in the session cache, a second session to the first potential service, wherein the second session comprises the second service parameter; and storing the second service parameter in the second session, wherein the first session and the second session share values of the first service parameter and the second service parameter.
7. The service recommendation method of claim 1 , wherein the first service corresponds to a second service of the plurality of services, and wherein a second degree of relevance between the second service and the first service indicates a probability of requesting the second service.
8. The service recommendation method of claim 1 , wherein the service recommendation method is executed by a server on a network side.
9. The service recommendation method of claim 1 , wherein the service recommendation method is executed by a terminal.
10. An apparatus with an intelligent assistant, comprising: a memory comprising instructions; and a processor coupled to the memory and configured to execute the instructions, which causes the processor to be configured to: receive a request for a first service by the intelligent assistant; parse the request for the first service using natural language processing; extract a service name from the request after parsing the request, wherein the service name indicates a type of the first service; determine a type of service parameter for the first service based on the service name, wherein the type of service parameter for the first service is recorded in a pre-established service relationship model, wherein the established service relationship model indicates a relationship between two or more services, wherein the pre-established service relationship model comprises a plurality of service names of a plurality of services provided by the intelligent assistant, corresponding degrees of relevance of the plurality of services to the first service, a plurality of corresponding trigger conditions of the plurality of services, and types of service parameters of the plurality of services, wherein each service parameter is a configuration parameter of a third-party application that provides a corresponding service, wherein each service parameter comprises a mandatory parameter and an optional parameter, wherein the mandatory parameter is a parameter that must be received to invoke the first potential service, and wherein the optional parameter is a parameter that is optionally received with the mandatory parameter; extract a service parameter for the first service from the request based on the type of service parameter indicated in the pre-established service relationship model after parsing the request; select a first potential service with a first degree of relevance to the first service based on the service name, the service parameter, and the pre-established service relationship model indicating that a degree of relevance between the first potential service and the first service meets a threshold; generate an instruction for requesting one of the third-party applications to provide the first potential service; detect whether a corresponding trigger condition for performing the first potential service, as indicated in the pre-established service relationship model, is met after generating the instruction; and determine whether to buffer the instruction in a cache or send the instruction to the one of the third-party applications based on whether the corresponding trigger condition is met.
11. The apparatus of claim 10 , wherein the threshold comprises at least one of a relevance degree or a quantity of recommended services.
12. The apparatus of claim 10 , wherein the threshold is for a quantity of services recommended to the user, and wherein the processor is further configured to: rank the corresponding degrees of relevance of the plurality of services to the first service; and select the first potential service corresponding to a ranking that meets the threshold.
13. The apparatus of claim 12 , wherein the processor is further configured to: select, from the pre-established service relationship model, a second potential service with a second degree of relevance to the first service that meets the corresponding trigger condition, wherein the first degree of relevance is higher than the second degree of relevance; and recommend, according to a ranking of the first degree of relevance and the second degree of relevance, the first potential service.
14. The apparatus of claim 13 , wherein the pre-established service relationship model further comprises corresponding relevance conditions of the plurality of services, and wherein the processor is further configured to: determine, a first relevance condition based on the request for the first service and an environmental condition of the first request; and select the first potential service when the first relevance condition meets the corresponding trigger condition.
15. The apparatus of claim 14 , wherein the processor is further configured to: extract, from the pre-established service relationship model, a first of the corresponding service parameters corresponding to the first service; extract, from the pre-established service relationship model, a second of the corresponding service parameters corresponding to the first potential service; allocate, in a session cache, a first session to the first service; store the first service parameter in the first session; allocate, in the session cache, a second session to the first potential service; and store the second service parameter in the second session, wherein the first session and the second session share values of the first service parameter and the second service parameter.
16. The apparatus of claim 15 , wherein the first service corresponds to a second service of the plurality of services, and wherein a second degree of relevance between the second service and the first service indicates a probability of requesting the second service.
17. The apparatus of claim 16 , wherein the apparatus is a server on a network side.
18. The apparatus of claim 16 , wherein the apparatus is a terminal.
19. The service recommendation method of claim 1 , wherein the first potential service is further selected based on a probability of the first potential service being selected by the user.
20. The apparatus of claim 10 , wherein the processor is further configured to select the first potential service based on a probability of the first potential service being selected by the user.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
December 28, 2015
March 16, 2021
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.